Why are my prompt_tokens higher than my text?

Two reasons:

1. A short Clatter Labs instruction block is added to every request and counts toward input tokens. The docs say so on the Model API overview.
2. Message formatting (roles, tool definitions, earlier turns) is tokenised too.

On a tiny message the difference looks big (the quickstart example shows 148 prompt tokens for a one-line haiku request). On real prompts it's a small share. usage always shows exactly what's billed.

Good to know#

• Your key's account is responsible for your end users' text under the Abuse policy. What it means for apps
• Every call shows in Console › Logs and Usage as Model API, with its price. See what you spent
• Every response has an X-Request-Id; log it so a request can be found later. Why log it

Still need help?#

• General support: support@clatterlabs.com or team@clatterlabs.xyz
• Infrastructure questions, or permanent deletion of your data (your whole account or one service): infrastructure@clatterlabs.xyz

We usually reply within the first hour, and always within 24 hours. A full support team is with you.